Large-Scale Survey for Tickborne Bacteria, Khammouan Province, Laos

Abstract

We screened 768 tick pools containing 6,962 ticks from Khammouan Province, Laos, by using quantitative real-time PCR and identified Rickettsia spp., Ehrlichia spp., and Borrelia spp. Sequencing of Rickettsia spp.–positive and Borrelia spp.–positive pools provided evidence for distinct genotypes. Our results identified bacteria with human disease potential in ticks in Laos.

Rickettsia, Borrelia, Ehrlichia, Anaplasma, and Coxiella spp. are tick-associated bacteria and well-described human pathogens. All of these bacteria, except Coxiella spp., are primarily transmitted through tick bites and cause febrile disease with a wide spectrum of severity. Tickborne bacterial pathogens are believed to be an underrecognized cause of acute febrile illness in Southeast Asia (1).

In Laos, spotted fever group Rickettsia have been shown to cause undifferentiated fever in 2% of febrile hospitalized adult patients (2). However, data on bacteria in ticks in Laos are sparse. To date, 1 Rickettsia sp. has been identified in a Boophilus sp. tick from Luang Namtha Province; this species showed 99.8% similarity with the Rickettsia sp. FUJ98 ompA gene (3). No other tickborne bacteria have been reported from Laos. Therefore, we investigated Rickettsia, Borrelia, Ehrlichia, Anaplasma, and Coxiella spp. in ticks from Khammouan Province, Laos.

The Study

We collected ticks in Nakai District, Khammouan Province, during the dry seasons (December–April) during 2012–2014, as previously described (4) (Technical Appendix Figures 1, 2). A total of 6,692 ticks were pooled (n = 768 pools, 1–10 ticks/pool) according to genus, sex, developmental stage, collection period, and site. One Amblyomma testudinarium nymph that contained a blood meal was processed separately.

We extracted DNA by using the NucleoSpin 8 Virus Extraction Kit (Macherey-Nagel, Düren, Germany). Pools were screened by using single quantitative real-time PCRs specific for Rickettsia spp. (17-kDa gene), Borrelia spp. (23S rRNA gene), Anaplasma spp. (major surface protein 2 gene), Ehrlichia spp. (16S rRNA gene), and Coxiella spp. (IS1111) (5–8) (Technical Appendix Table 1). Five microliters of diluted (1:10) template containing 1× Platinum Supermix-UDG (Invitrogen, Carlsbad, CA, USA) and bovine serum albumin (40 mg/mL) were used for each assay. Positive and nontemplate controls were included in each run. Screening by PCR was performed once per sample. In concordance with published guidelines, results were considered positive if they had a cycle quantitation (C_q) value <40 and likely positive if they had a C_q value 40–45 (9).

Sequencing was attempted for pools with C_q values <40 (online Technical Appendix Table 2) and performed by Macrogen (Seoul, South Korea). Consensus sequences were analyzed by using CLC Main Workbench 7 (http://www.clcbio.com/products/clc-main-workbench/) and BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and submitted to GenBank. Phylogenetic trees were constructed by using the Kimura-2 parameter model and the neighbor-joining method. Bootstrap values were determined by using 1,000 replications.

Table 2. Sequence data for Rickettsia species isolated from ticks, Khammouan Province, Laos*.

Tick pool	Tick species and stage	Rickettsia spp. gene, GenBank accession no., and % similarity (no. matching nucleotides/total)
		17-kDa	gltA	sca4	ompA	ompB
110	Amblyomma testudinarium nymph	Unclear sequence	NS	Unclear sequence	KT753264, 100.0 (529/529) with Rickettsia sp. TwKM01 EF219467	NS
177, 180, 216, 220	A. testudinarium nymph	KR733070, 100.0 (355/355) with R. tamurae AB114825	KT753265, 99.8 (1,096/1,098) with R. tamurae AB812551	KT753266, 99.7 (607/609) with R. tamurae DQ113911	NS	NS
315	A. testudinarium nymph	KT753267, 98.8 (407/412) with R. raoultii JX885457	KT753268, 99.9 (1,036/1,037) with Ricksettia kagoshima6 JQ697956	KT753269, 96.8 (795/821) with Rickettsia sp. AUS 118, KF666473	Could not be amplified	KT753270, 95.0 (1,073/1,129) with R. massiliae CP003319
239	A. testudinarium nymph	KT753271, 99.7 (360/361) with Rickettsia sp. ATT AF483196	KT753272, 99.7 (1,048/1,051) with R. tamurae (AB812551)/KT753273; 99.2 (367/370) with Rickettsia sp. hhmj7 KC566999	KT753274, 97.1 (759/782) with Rickettsia sp. AUS 118 KF666473	KT753275, 87.2 (530/602) with R. raoultii JQ792137	KT753276, 97.5 (1,052/1,079) with R. massiliae CP003319
76, 337, 450, 453	Haemaphysalis G1 nymphs (3), A. testudinarium nymph (1)	KT753277, 98.4 (417/423) with R. raoultii JX885457	KT753278, 99.9 (1,037/1,038) with Ricksettia sp. kagoshima6 JQ697956	KT753279, 98.4 (794/807) with R. japonica AF155055	Could not be amplified	KT753280, 96.0 (410/427) with R. raoultii EU036984
81, 372	Haemaphysalis G1 nymphs, A. testudinarium nymph (17 kDa only)	KT753283, 99.0 (408/412) with R. raoultii JX885457	KT753284, 99.5 (1,090/1,096) with R. sibirica U59734	KT753285, 98.5 (838/851) with R. japonica AF 155055	Could not be amplified	KT753286, 97.7 (1,118/1,144) with R. massiliae CP003319
120	Haemaphysalis G1 nymph	KT753287, 96.1 (391/407) with R. helvetica GU827073	KT753288, 97.1 (370/381) with Candidatus Rickettsia rara DQ365805	Could not be amplified (x2)	Could not be amplified (x2)	KT753289, 86.4 (362/419), R. aeschlimannii AF123705
407	Haemaphysalis hysticis adult	KR733074, 100.0 (413/413), R. japonica AP011533	KT753281, 100.0 (1,063/1,063), R. japonica AP011533	NS	NS	KT753282, 100.0 (1,191/1,191) with R. japonica AP011533
447	Haemaphysalis G1 nymph	KT753291, 98.6 (407/413) with R. massiliae CP000683	KT753290, 99.6 (961/965) with R. raoultii JX885455	KT753292, 97.5 (809/830) with Rickettsia sp. AUS118 KF66473	KT753293, 97.5 (591/606) with Rickettsia sp. JL-02 AY093696	KT753294, 98.4 (1,137/1,156), with R. massiliae CP003319

*New sequences were compared with reference sequences. NS, not sequenced.

A total of 768 tick pools containing 6,692 ticks were screened. Pools contained 3 genera of ticks: 59.9% (460/768) Haemaphysalis spp., 36.3% (279/768) A. testudinarium, and 3.8% (29/768) Dermacentor auratus. Of the pools, 3% (23/768) contained adults, 36.5% (280/768) contained larvae, and 60.5% (465/768) contained nymphs (Table 1).

Table 1. Tick pools tested for bacteria after screening by quantitative PCR, Khammouan Province, Laos*.

Bacteria and tick species	No. positive pools/no. tested (%)
	Total	Larvae	Nymphs	Adult males	Adult females
Rickettsia spp.
All	44/768 (5.7)	6/280 (2.1)	37/465 (8.2)	0/12 (0)	1/11 (9.1)
Amblyomma testudinarium	27/279 (10.0)	0/61 (0)	27/217 (12.9)	0/1 (0)	0/1 (0)
Haemaphysalis G1	5/398 (3.8)	6/194 (3.1)	9/200 (4.5)	0/3 (0)	0/1 (0)
H. hystricis	1/6 (16.7)	NS	NS	0/3 (0)	1/3 (33.3)
Dermacentor auratus	1/29 (3.4)	0/0 (0)	1/26 (3.8)	0/2 (0)	0/1 (0)
Ehrlichia spp.
All	12/768 (1.6)	4/280 (1.4)	6/465 (1.3)	1/12 (8.3)	1/11 (9.1)
A. testudinarium	2/279 (0.7)	0/61 (0)	2/217 (0.9)	0/1 (0)	0/1 (0)
Haemaphysalis G1	8/398 (2.0)	4/194 (2.1)	4/200 (2.0)	0/3 (0)	0/1 (0)
H. aborensis	2/6 (33.3)	NS	NS	1/3 (33.3)	1/3 (33.3)
Borrelia spp.
All	12/768 (1.6)	2/280 (0.7)	8/465 (1.7)	2/12 (16.7)	NS
A. testudinarium	2/279 (0.7)	1/61 (1.6)	1/217 (0.5)	0/1 (0)	0/1 (0)
Haemaphysalis G1	6/398 (1.5)	1/194 (0.5)	5/200 (2.5)	0/3 (0)	0/1 (0)
Haemaphyalis G1.2	1/13 (7.7)	NS	1/13 (7.7)	NS	NS
H. aborensis	2/6 (33.3)	NS	NS	2/3 (66.7)	0/3 (0)
D. auratus	1/29 (3.4)	0/0 (0)	1/26 (3.8)	0/2 (0)	0/1 (0)
Coxiella spp.
All	5/511 (1.0)†	4/187 (2.1)†	1/310 (0.3)	0/8 (0)	0/6 (0)
Haemaphysalis G1	5/279 (1.8)†	4/162 (2.5)†	1/117 (0.9)	NS	NS
Anaplasma spp.
All	2/768 (0.3)†	0/280 (0)†	0/465 (0)†	0/12 (0)	0/11 (0)
A. testudinarium	1/279 (0.4)†	0/61 (0)	1/217 (0.5)†	0/1 (0)	0/1 (0)
Haemaphysalis G1	1/398 (0.3)†	1/194 (0.5)†	0/200 (0)	0/3 (0)	0/1 (0)

*NS, no samples were available for screening.
†Includes samples with cycle quantitation values <40 and 40–45.

Rickettsia spp. were identified in 5.7% (44/768) of pools, and an additional 2.3% (18/768) of pools were likely positive for Rickettsia spp. Sequences consistent with 5 described Rickettsia species or genotypes were identified: R. tamurae, R. japonica, Rickettsia sp. ATT, Rickettsia sp. Kagoshima6, and Rickettsia sp. TwKM01 (Table 2; Figure 1).

Figure 1.

Phylogenetic analysis of Rickettsia spp. in ticks, Khammouan Province, Laos. The tree was constructed by using partial nucleotide sequences (350 bp) of the 17-kDa gene, the Kimura 2-parameter model, and the neighbor-joining method. Analyses were supported by bootstrap analysis with 1,000 replications. Numbers along branches are bootstrap values. GenBank accession numbers are shown for reference sequences. Sample numbers for each tick are shown in parentheses. Scale bar indicates nucleotide substitutions per site.

Three novel genotypes (Table 2) were identified that might be new species. Candidatus Rickettsia laoensis (pool 447) was identified in 1 Haemaphysalis sp. pool. Phylogenetic analysis of 2845–2920-bp concatenated sequences of gltA, sca4, and ompB genes suggested that this bacteria belonged to the R. massiliae group of rickettsiae (Technical Appendix Figure 3). Candidatus Rickettsia mahosotii (pools 81 and 372) was identified in Haemaphysalis spp. and A. testudinarium pools. Phylogenetic analysis of gltA, sca4, and ompB genes suggested that this bacteria belonged to the R. rickettsii group (Technical Appendix Figure 3). Candidatus Rickettsia khammouanensis was identified in 1 Haemaphysalis sp. nymph pool (pool 120). Phylogenetic analysis of gltA, 17-kDa, and ompB genes suggested a relationship with the R. helvetica group (Technical Appendix Figure 4).

In addition, 15 A. testudinarium pools showed dual peaks for 17-kDa gene sequences, which suggested the presence of R. tamurae and Rickettsia sp. ATT. Sequencing of sca4, ompA, and ompB genes from 1 of these pools (pool 239) identified unique sequences (Table 2; Technical Appendix Figure 4).

Borrelia spp. were identified in 1.6% (12/768) of pools (Table 1). Two unique sequences obtained from Haemaphysalis spp. pools showed 99.3% (298/300) (GenBank accession no. KR733069) and 98.7% (296/300) (accession no. KR733068) identity with Shiretoko Haemaphysalis Borrelia sp. (AB897888). Phylogenetic analysis confirmed that both bacteria were closely related to Shiretoko Haemaphysalis Borrelia sp. (accession no. B897888) and belong to the relapsing fever group of Borrelia (Figure 2).

Figure 2.

Phylogenetic analysis of Borrelia spp. in ticks, Khammouan Province, Laos. The tree was constructed by using partial nucleotide sequences (299–323 bp) of the flaB gene, the Kimura 2-parameter model, and the neighbor-joining method. Analyses were supported by bootstrap analysis with 1,000 replications. Numbers along branches are bootstrap values. GenBank accession numbers are shown for reference sequences. Sample numbers for each tick are shown in parentheses. Scale bar indicates nucleotide substitutions per site.

Twelve (1.6%) of 768 pools were positive for Ehrlichia spp. (Table 1); an additional 6 pools (0.8%) were likely positive. One short sequence from a Haemaphysalis sp. nymph pool (pool 357) was obtained, and this sequence showed 100% identity (116/116 bases) with the genus Ehrlichia.

No pools were positive for Anaplasma spp., but 2 were likely positive (Table 1). Although not all pools were tested for Coxiella spp. (n = 511), 1 pool (0.2%) was positive, and 4 pools were likely positive for C. burnetti. No confirmatory sequences were obtained from these pools. The 1 tick that contained a blood meal (A. testudinarium nymph) showed negative results by screening PCRs.

Conclusions

This study provides evidence that Rickettsia spp., Borrelia spp., and Ehrlichia spp. are present in ticks in Laos. Several Rickettsia spp. identified in this study are human pathogens. Infections with R. tamurae (2) and R. japonica are well described in Southeast Asia (10). However, the pathogenicity of Rickettsia sp. TwkM01 (11), Rickettsia sp. ATT (12), Rickettsia sp. kagoshima6 genotypes (13) and potential novel Candidatus Rickettsia laoensis, Candidatus Rickettsia mahosotii, and Candidatus Rickettsia khammouanensis is unknown. Candidatus Rickettsia khammouanensis is phylogenetically related to R. helvetica, for which there is serologic evidence for its role as a human pathogen in Laos (2). Unique ompA, ompB, and sca4 sequences identified in this study (Table 2) might indicate the presence of Rickettsia sp. ATT (12), which was previously believed to be identical to R. tamurae (14), and suggests that it might be a distinct species. Further studies, including whole-genome sequencing, are required to identify and confirm these novel genotypes and understand their role in human disease.

Borrelia spp. sequences identified in Haemaphysalis spp. pools were shown to have high concordance with the Shiretoko Haemaphysalis Borrelia isolated from Haemaphysalis spp. ticks and deer in Japan (15). The species belongs to the relapsing fever group of Borrelia and is related to B. lonestari.

Sequence data for Ehrlichia spp. indicated the presence of these bacteria but were not sufficient to identify them to the species level. The C_q values were high (40–45) for Anaplasma spp., but no sequence data were obtained. Coxiella spp. were screened by using primers for IS1111, which are not specific for C. burnetii, and no confirmatory sequence data were obtained. Because of limited reagents, screening of all 768 pools for Coxiella spp. was not completed. Further work is required to investigate the presence of these bacteria in Laos.

Our study had several limitations. First, pooling of ticks precludes an accurate assessment of prevalence of bacterial pathogens. Second, sequences obtained from some A. testudinarium pools had dual peaks, suggestive of multiple infections, and could therefore not be interpreted. Third, ticks were collected only from 1 area in Laos (Khammouan Province); thus, extrapolating findings to the entire country must be done cautiously.

Our results highlight the frequency of tickborne bacterial infections in Laos. These findings emphasize the need for further research of tick-associated bacteria and their role in human disease.

Technical Appendix

Additional information on large-scale survey for tickborne bacteria, Khammouan Province, Laos.

Biography

Dr. Taylor is a research physician at the Centre for Tropical Medicine and Global Health, University of Oxford, Oxford, UK. His primary research interest is infectious diseases.